麻豆传媒

Neoadjuvant chemoradiation failed to improve survival in borderline-resectable pancreatic cancer as compared with chemotherapy alone, a randomized trial showed.

Patients randomized to chemotherapy alone had an 18-month OS rate of 66.4% versus 47.3% for the group that received chemoradiation (CRT). A futility analysis showed that adding stereotactic body radiation therapy (SBRT) was unlikely to improve overall survival (OS).

"Preoperative modified FOLFIRINOX was associated with favorable overall survival relative to historical data in patients with borderline-resectable pancreatic ductal adenocarcinoma," Matthew H.G. Katz, MD, of the University of Texas MD Anderson Cancer Center in Houston, said during the Gastrointestinal Cancers Symposium virtual meeting. "We cannot conclude efficacy for modified FOLFIRINOX plus radiation, as statistical requirements were not met.

"Based on the results of this study, modified FOLFIRINOX represents a reference preoperative regimen for patients with borderline-resectable pancreatic adenocarcinoma," he said.

The trial was not designed or statistically powered for a direct comparison of preoperative chemotherapy with or without radiation therapy, said invited discussant Rebecca Snyder, MD, of the Brody College of Medicine at East Carolina University in Greenville, North Carolina. Nonetheless, the results extended a procession of unimpressive data for radiation therapy in pancreatic cancer.

"There is no convincing randomized data that radiation therapy prolongs survival in any population of unselected patients with pancreatic cancer," said Snyder. "Certainly, follow-up questions remain, specifically regarding the role for non-SBRT radiation in the preoperative setting, which may be answered in the ongoing and trials."

"The addition of SBRT for the management of borderline disease in the preoperative setting does not appear to be justified," she continued. "The putative survival benefit of preoperative conventional radiation, specifically in combination with modified FOLFIRINOX, remains unknown, although based on the existing data, a significant survival benefit seems unlikely."

The results do not rule out a benefit for certain subsets of high-risk patients, but those groups have yet to be identified, Snyder added.

In reviewing the background of the study, Katz noted that the American Society of Clinical Oncology recommends preoperative therapy for "localized pancreatic adenocarcinoma who have tumors that have a significant radiographic interface with a major mesenteric blood vessels, as these patients are at high risk for a margin-positive operation and short survival when pancreatectomy is performed de novo."

Both chemotherapy and radiotherapy are often used in the setting of so-called borderline-resectable pancreatic cancers, but the optimal regimen remains controversial, he continued. The was designed to identify a reference neoadjuvant regimen for future studies in pancreatic cancer.

Patients who met radiographic criteria for borderline-resectable pancreatic cancer were randomized to eight cycles of FOLFIRINOX chemotherapy or to seven cycles of FOLFIRINOX plus radiation therapy with SBRT or hypofractionated image-guided radiotherapy. Following restaging, patients suitable for surgery underwent pancreatectomy and then received four cycles of adjuvant FOLFOX chemotherapy.

The primary endpoint was 18-month OS, and key secondary endpoints included event-free survival (EFS), adverse events (AEs), clear surgical margins (R0), and pathologic complete response (pCR) rate. If either arm met full accrual (62 patients) and at least 36 patients were alive at 18 months, the regimen was considered efficacious. If both arms were efficacious, the regimen associated with better results would be declared the winner.

An interim analysis of the first 30 patients showed that 17 (57%) had R0 resection in the chemotherapy arm versus 10 (33%) in the CRT arm. When the study ended, 70 patients had been enrolled in the chemotherapy arm and 56 in the CRT arm. Katz reported that 20 patients (31%) completed all planned therapy in the chemotherapy arm as compared with 10 (18%) in the chemoradiation group.

The 65 evaluable patients in the chemotherapy arm received a median of eight cycles of FOLFIRINOX, and the 55 patients in the CRT arm received a median of seven cycles of FOLFIRINOX. Dose reductions and treatment delay occurred slightly more often in the CRT arm (75% vs 60% and 60% vs 49%, respectively). Grade 3/4 chemotherapy-related AEs occurred in a similar proportion of patients in each group.

Katz reported that 32 patients (49%) in the chemotherapy arm and 19 (35%) in the CRT group underwent surgery, which resulted in R0 status in 88% and 74% of the respective groups. Two pCRs occurred in the CRT arm versus none in the chemotherapy arm.

Data on surgical AEs showed that anemia, fistula/abscess, wound infection, and re-admission occurred more often in the CRT group.

Survival data for 65 patients in each arm showed a median OS of 29.8 months with chemotherapy alone versus 17.1 months for the CRT arm. Median EFS was 15.0 versus 10.2 months.

Comment